AB-410 Practice Test 19 — 15 Questions

Question 1

What happens when you delete a 'Managed' solution in a target environment?

Accepted Answer

The components are removed. Deleting a managed solution from a target environment removes the solution and all components it owns, provided no other solution layer depends on them. Managed solutions are deployed as immutable layers, and uninstalling the managed solution deletes the components that were introduced by that solution. Data stored in those tables is not deleted by default.

Answer

The data is deleted

Answer

The solution becomes unmanaged

Answer

Nothing happens

Question 2

Which component allows users to navigate between different entities in a Model-driven app?

Accepted Answer

Site Map. The Site Map defines the navigation structure of a model-driven app, including areas, groups, and subareas that link to entities, dashboards, and other components. It is the component that allows users to navigate between different entities. Without a site map, users cannot move between tables in the app.

Answer

Business Process Flow

Answer

View

Answer

Form

Question 3

A team is building a canvas app that calls a Power Automate cloud flow to create a record in Dataverse when a user selects a button. The flow runs but returns no data back to the app, and the app's label never updates. You confirm the flow completes successfully and returns a response. What should you verify in the app to ensure the returned value is displayed?

Accepted Answer

That the label's Text property references the flow's response output. A canvas app retrieves values from a Power Automate flow through the flow's output object, and any control that should display that data must bind to it. Because the flow already runs and returns a response, the remaining step is ensuring the label's Text property references that response output. The other checks concern flow triggering, direct Dataverse access, or local collections, none of which affect displaying the returned value.

Answer

That the flow's Run only when condition is set to true

Answer

That the button's OnSelect property uses a Collect function

Answer

That the app has the Dataverse connector added

Question 4

You are creating a new custom table in Dataverse to store product information. You need to ensure that each product has a unique identifier that is automatically generated and can be used as a primary key. Which column type should you use?

Accepted Answer

Autonumber. An Autonumber column is designed to automatically generate unique numeric values for each record. It can be customized with prefixes and suffixes, making it suitable for use as a primary key. This ensures that every product has a distinct identifier without manual data entry, simplifying record management and integration.

Answer

Date and Time

Answer

Currency

Answer

Text

Question 5

A developer is creating a custom connector for a REST API that requires an API key in the header. The connector will be used in a canvas app. The developer wants to ensure that the API key is not exposed to users of the app. What should the developer do?

Accepted Answer

Configure the connector's security to use API Key authentication.. The secure way to use an API key in a custom connector is to configure the connector's security settings to use API Key authentication. The key is then stored securely and injected by the connector at runtime, so app users never see it. Hardcoding, using environment variables, or asking users for the key are insecure and expose the secret.

Answer

Ask users to enter the API key in the app, and store it in a collection.

Answer

Hardcode the API key in the connector's Swagger definition.

Answer

Store the API key in a Power Apps environment variable and reference it in the connector.

Question 6

You are building a Power Automate cloud flow that must call an Azure Function. The Azure Function requires authentication and returns a JSON payload. You need to pass a secret key to the function securely. Where should you store the secret key?

Accepted Answer

In Azure Key Vault, and retrieve it at runtime using the 'Get secret' action in Power Automate.. Azure Key Vault is the correct place to store secrets. Power Automate's Key Vault connector includes a 'Get secret' action that retrieves the secret securely at runtime. The secret is not exposed in the flow definition or run history, and access is governed by Azure role-based access control, ensuring only authorized flows can retrieve it.

Answer

In a Microsoft Dataverse table with column-level security enabled, and retrieve it with a 'List rows' action.

Answer

In the flow's definition as a plain text value in the Azure Function action's headers.

Answer

In a flow variable defined with the 'Initialize variable' action, marked as secure.

Question 7

You are developing a canvas app that uses a custom connector to call an Azure Function. The function requires an API key. You need to store the API key securely so that it is not exposed in the app's source code or to users. What should you do?

Accepted Answer

Configure the custom connector to use API key authentication and enter the key in the connector's security settings.. The secure way to handle an API key for a custom connector is to configure API key authentication in the connector's security settings. The key is stored encrypted and not exposed in the app. Storing it in variables, SharePoint lists, or hardcoding it in the definition file all risk exposure and do not meet the security requirement.

Answer

Hardcode the API key directly in the custom connector's definition file and import it into the environment.

Answer

Store the API key in a SharePoint list and retrieve it at runtime using a LookUp function.

Answer

Store the API key in a variable using Set(varApiKey, "...") in the App.OnStart property.

Question 8

You are building a canvas app that calls a custom connector to an internal REST API. The API requires an API key that must never be exposed to app users or stored in the app itself. You need the connector to authenticate without embedding the secret in the app. What should you configure?

Accepted Answer

Configure the custom connector to use API key authentication, then create a connection that stores the key in the connection credentials.. The requirement is to authenticate a custom connector without exposing the API key in the app. Custom connectors support API key authentication where the key is entered when creating the connection, and the platform stores it securely with the connection. The app then uses the connection, so the secret never appears in the app definition or to users.

Answer

Enable the connector's OAuth 2.0 implicit grant flow and pass the API key as a scope parameter.

Answer

Store the API key in a variable initialized in the App.OnStart property.

Answer

Add the API key as a text input control on a hidden screen and reference it in the connector call.

Question 9

A canvas app uses a Microsoft Copilot Studio agent to summarize long text fields. Users report that the first summary request after the app has been idle for several minutes takes much longer than subsequent requests, and occasionally times out. You must reduce the perceived latency without changing what the agent returns. What should you do?

Accepted Answer

Send a lightweight warm-up request to the agent when the relevant screen becomes visible.. The delay is a cold-start effect on the agent, so priming it with a lightweight request when the screen appears lets initialization happen in the background while the user works. The real summarize call then runs against a warm agent, cutting perceived latency and reducing timeouts, without changing the agent's response content.

Answer

Display a progress indicator and disable the summarize button while the agent call is in flight.

Answer

Move the agent call into an OnStart expression so it runs when the app loads rather than on first use.

Answer

Increase the app's data row limit so the agent has more context available on the first call.

Question 10

A canvas app calls a Microsoft Copilot Studio agent that returns a long generated summary. Users on tablets report the summary is cut off in a label, and scrolling within the label is not possible. You must show the full summary while keeping the layout stable. What should you do?

Accepted Answer

Place the label inside a vertical container with scrolling enabled and set the label to auto-height.. To show a long summary without destabilizing the layout, put the auto-height label inside a vertical container that has scrolling enabled. The label expands to its full content height, and the container constrains the visible area while letting users scroll to read everything, which satisfies both requirements on tablets.

Answer

Set the label's Text property to a fixed number of characters and append an ellipsis.

Answer

Reduce the label's font size so the full summary fits within the fixed label height.

Answer

Set the label's AutoHeight property to true and place it inside a vertical container.

Question 11

Which THREE of the following can cause an AI model to fail in production? (Choose THREE)

Accepted Answer

Changes in input data format. Option A is correct because a production model expects inputs in the exact schema, encoding, and feature order used during training; if the input data format changes (e.g., a field switches from numeric to string, timestamps change timezone, or JSON keys are renamed), preprocessing or inference fails or produces garbage predictions. Option B is correct because expired service account credentials (e.g., an OAuth 2.0 token or API key past its expiry) cause authentication failures such as HTTP 401 Unauthorized, so the model endpoint cannot be called and the AI service effectively fails. Option E is correct because insufficient API service quota (e.g., requests-per-minute or tokens-per-minute limits on the model endpoint) leads to HTTP 429 Too Many Requests or throttling, so production traffic is rejected even though the model itself is healthy. Option C is not correct because increasing flow speed is a networking/throughput characteristic and does not by itself break model inference. Option D is not correct because changing a flow name is just a label change and does not affect the model's ability to serve predictions.

Answer

Increasing flow speed

Answer

Changing the flow name

Question 12

You want to automate the extraction of data from invoices using AI Builder. Which model type should you choose?

Accepted Answer

Document processing model. Document processing models in AI Builder are designed to extract structured data from documents like invoices, receipts, and forms. They combine OCR with layout understanding to identify fields such as invoice number, date, and total. This is the correct model type for automating invoice data extraction.

Answer

Object detection model

Answer

Category classification model

Answer

Prediction model

Question 13

A system administrator is setting up a new Dataverse environment. They need to ensure that users can only access records that they own or that are shared with them, and that they cannot see records owned by other users in the same business unit. Which security concept should they configure?

Accepted Answer

Security roles with User-level access. Security roles in Dataverse define both privileges (what actions a user can perform) and access levels (which records they can perform them on). By assigning a security role with 'User' access level for the relevant entities, users can only access records they own or that are explicitly shared with them. This prevents them from seeing records owned by other users, even within the same business unit, unless sharing is configured.

Answer

Column security profiles

Answer

Field-level security

Answer

Business unit hierarchy

Question 14

You are building a canvas app that connects to Dataverse. You need to store a list of product categories that will be used across multiple apps. The categories should be easy to update without modifying any app. What should you create?

Accepted Answer

A custom table named Category with a primary column for the category name.. A custom table in Dataverse is the appropriate choice for storing shared reference data that multiple apps can consume. It centralizes data management, supports security roles, and allows updates without app modifications. Other options are either app-specific or lack the necessary structure and manageability.

Answer

An Excel file stored in OneDrive and connected via the Excel connector.

Answer

A static collection defined in each app's OnStart property.

Answer

A global choice column in the Product table.

Question 15

Which TWO settings are crucial when deploying a model-driven app to production?

Accepted Answer

Environment variables. Environment variables (A) are crucial because they let you store environment-specific values such as connection strings, API endpoints, and SharePoint site URLs outside the app definition, so the same app can be promoted from dev to test to production without code changes. Managed solutions (E) are equally crucial because they are the supported deployment vehicle for production: they are locked against direct customization, support clean upgrade and patch paths, and allow clean uninstall, unlike unmanaged solutions. Hard-coded URLs (B) are wrong because they break when the app moves between environments and defeat the purpose of environment variables. Manual editing of production forms (C) is wrong because production customizations must flow through solutions, not ad-hoc edits that create unmanaged layers and drift. Local developer PC IP addresses (D) are wrong because they are machine-specific and irrelevant to a cloud-hosted model-driven app's production configuration.

Answer

Hard-coded URLs

Answer

Manual editing of production forms

Answer

Local developer PC IP addresses